Mutations at the adenosine deaminase (ADA) locus which diminish enzyme activity result in three major phenotypes. Complete absence of enzyme activity in all cell types results in a syndrome of Severe Combined Immunodeficiency (SCID) which is fatal if untreated. Absence of enzyme activity in RBCs with retention of variable amounts of residual enzyme activity in other cell types, ("partial" ADA deficiency) is usually associated with grossly normal immune function. However, "partial" ADA deficiency can also present as a late onset milder, slowly progressive immunodeficiency. We propose that the underlying genotype will most accurately predict eventual outcome in partially ADA deficients as well as response to different forms of therapy in ADA-SCIDs. Such therapy currently includes enzyme replacement and bone marrow candidate for gene therapy, it will be of prime importance to predict potential variation in the response to such therapy due to the nature of the basic mutation. We with various phenotypes and variables responses to therapy. We will use classic cDNA cloning and sequencing and/or amplification or genomic DNA by polymerase chain reaction (PCR) and direct sequencing of amplified DNA. We also propose that many ADA SCID's have small but significant residual enzyme activity not detectable by usual assays. We will unambiguously determine the heterologous cells under conditions sufficiently sensitive to detect very low amounts of human ADA activity. Although ADA is a ubiquitous enzyme, the steady state activity varies markedly in different tissues, during the cell cycle and during differentiation. We propose, based on preliminary data, that during differentiation a switch occurs in the sites but not transported efficiently to the cytoplasm and results in a decrease in steady state mRNA. We will determine rates and sites of initiation by nuclear run on and RNAase protection experiments and will determine amounts of steady state mRNA and ADA activity.